AI maps the brain’s hidden bridge and reveals genetic links to mental health

Summary: For the first time, scientists have mapped the genetic architecture of the brain’s communication bridge (the corpus callosum) using AI and MRI data from more than 50,000 people. The study discovered dozens of genes that shape the size and thickness of this vital structure, many of them active during prenatal development, when brain wiring is established.

Differences in these genes may explain why changes in the corpus callosum are associated with mental and neurological disorders. The team’s open source artificial intelligence tool now allows global researchers to analyze brain structure faster and with unprecedented precision.

Key facts

Genetic close-up: identified genetic regions that shape the brain’s left-right communication bridge. AI-powered discovery: Artificial intelligence was used to map brain structure in 50,000 MRI scans in record time. Mental health connection: Genetic overlap found between corpus callosum, cerebral cortex, ADHD and bipolar disorder.

Source: USC

For the first time, a research team led by the Mark and Mary Stevens Institute for Computing and Neuroimaging (Stevens INI) at the Keck School of Medicine of USC has mapped the genetic architecture of a crucial part of the human brain known as the corpus callosum, the thick band of nerve fibers that connects the left and right hemispheres of the brain.

The findings open new avenues for discoveries about mental illnesses, neurological disorders and other diseases related to defects in this part of the brain.

The corpus callosum is essential for almost everything the brain does, from coordinating the movement of our limbs in synchronization, to integrating images and sounds, to higher-order thinking and decision making.

Abnormalities in their shape and size have long been linked to disorders such as ADHD, bipolar disorder and Parkinson’s disease. Until now, the genetic underpinnings of this vital structure remained largely unknown.

In the new study, published in Nature Communications, the team analyzed brain scans and genetic data from more than 50,000 people, from childhood to late adulthood, with the help of a new tool the team created that leverages artificial intelligence.

“We developed an artificial intelligence tool that finds the corpus callosum in different types of brain MRIs and takes its measurements automatically,” said Shruti P. Gadewar, co-first author of the study and a research specialist at Stevens INI.

Using this tool, the researchers identified dozens of genetic regions that influence the size and thickness of the corpus callosum and its subregions.

“These findings provide a genetic model for one of the brain’s most essential communication pathways,” said Ravi R. Bhatt, PhD, co-first author of the study and a postdoctoral scholar in the Imaging Genetics Center at Stevens INI.

“By discovering how specific genes shape the corpus callosum and its subregions, we can begin to understand why differences in this structure are linked to various neurological and mental health conditions at the molecular level.”

The study revealed that different sets of genes govern the area versus thickness of the corpus callosum, two characteristics that change throughout life and play distinct roles in brain function. Several of the genes involved are active during prenatal brain development, particularly in processes such as cell growth, programmed cell death, and wiring of nerve fibers between hemispheres.

“This work demonstrates the power of using AI and large-scale databases to uncover the genetic factors that drive brain development,” said Neda Jahanshad, PhD, associate professor of neurology and senior author.

“By linking genetics with brain structure, we gain critical insight into the biological pathways that may underlie psychiatric and neurological diseases.”

In particular, the study found a genetic overlap between the corpus callosum and the cerebral cortex (the outer layer of the brain responsible for memory, attention and language), as well as with conditions such as ADHD and bipolar disorder.

“These connections underscore that the same genetic factors that shape the brain’s communication bridge may also contribute to vulnerability to certain disorders,” Jahanshad added.

Arthur W. Toga, PhD, director of the Stevens INI, emphasized the broader implications of this research. “This study is a milestone in understanding how our brain is built. Not only does it shed light on normal brain development, it also helps us identify new avenues to diagnose and potentially treat disorders that affect millions of people around the world.”

Researchers have made their new AI-based tool available to the public to accelerate future discoveries. The software, developed at Stevens INI, uses advanced machine learning to automatically identify and measure the corpus callosum from MRI scans. This approach allows scientists to analyze brain structure at an unprecedented scale and level of precision, reducing years of manual work to just hours.

Stevens INI has become a world leader in the application of artificial intelligence to neuroscience, developing tools that are freely shared with the research community. By combining massive data sets with cutting-edge computational methods, the institute is transforming the way scientists study brain health and disease.

“Artificial intelligence is revolutionizing brain research and Stevens INI is at the forefront of that revolution,” Toga said. “By pioneering artificial intelligence tools and making them widely available, we are empowering scientists around the world to unlock new discoveries about the brain much faster than ever before.”

About the study

In addition to Bhatt, Gadewar and Jahanshad, the study’s other authors include Ankush Shetty, Iyad Ba Gari, Elizabeth Haddad, Shayan Javid, Abhinaav Ramesh, Elnaz Nourollahimoghadam, Alyssa H. Zhu, Christiaan de Leeuw, Paul M. Thompson and Sarah E. Medland.

Funding: This work was supported by the National Institutes of Health (Grant Nos. R01 MH134004 and R01 AG059874 (NJ), National Science Foundation Graduate Research Fellowship Program (Grant No. 2020290241 (RRB), R01 MH126213, R01NS105746, the Cognitive Development Study Adolescent Brain Research (ABCD) (https://abcdstudy.org) and UK Biobank (Resource Application No. 11559 was supported by NHMRC grants APP1172917 and APP1158127).

Key questions answered:

About this news on AI, genetics and mental health research

Author: Laura LeBlanc
Source: USC
Contact: Laura LeBlanc – USC
Image: Image is credited to Neuroscience News.

Original research: Open access.
“The genetic architecture of the human corpus callosum and its subregions” by Ravi R. Bhatt et al. Nature Communications

Abstract

The genetic architecture of the human corpus callosum and its subregions

The corpus callosum (CC) is the largest set of white matter fibers that connect the two hemispheres of the brain. In humans, it is essential for coordinating sensorimotor responses and performing associative or executive functions.

Identifying which genetic variants underlie CC morphometry may provide molecular insights into the role of CC in mediating cognitive processes.

We developed and used an artificial intelligence-based tool to extract the total and regional area and thickness of the mid-sagittal CC in two large public data sets.

We performed a genome-wide association study (GWAS) meta-analysis of European participants (pooled N = 46,685) with generalization to non-European participants (pooled N = 7,040). Post-GWAS analyzes implicated prenatal intracellular organization and cell growth patterns, and high heritability in open chromatin regions.

The results suggest that immune-mediated programmed cell death drives thinning of the posterior body and isthmus.

Genetic overlaps and causal genetic liability were identified between CHD, features of the cerebral cortex, and neuropsychiatric disorders such as attention deficit/hyperactivity disorder, bipolar disorders, and Parkinson’s disease.